Does training more frequently lead to bigger strength gains (part two)?
The effect of training frequency on strength gains is difficult to assess. In the fitness industry, there are strong proponents of both infrequent (once per week) and very frequent (6+ times per week) training approaches, both on a body-part and on a full-body basis. In the literature, there are a number of studies but many of them do not control for the effect of increased volume. This review sets out what we currently know about how frequency affects strength gains, where volume is maintained the same.
What is the background?
Training frequency has traditionally been considered important for strength gains. However, training frequency is often (but not always) manipulated for the purposes of indirectly altering total weekly training volume. Indeed, in many research studies investigating frequency, total weekly training volume is often not equated between the groups. This leads to a greater total volume of training being performed by the high-frequency group. Since volume may also be a key factor, this is a confounding factor.
Therefore, it is important to consider what happens to strength gains when frequency is altered while maintaining total weekly training volume the same. This will provide information about whether splitting the same total weekly workload into more sessions would be superior to performing fewer but longer training sessions.
The purpose of this short review is to assess the effects of training frequency on strength gains measured by any metric in volume-matched studies of resistance-training-only interventions in both trained and untrained populations, where training frequency is >1 session per week. This involves the following selection criteria:
Including any intervention assessing the effects of training frequency on strength gains.
Measurement of strength gains by any metric (e.g. dynamic/isoinertial, isometric or isokinetic).
Excluding interventions that do not control for total weekly training volume.
Excluding interventions with aerobic exercise or other components that are not resistance-training.
Excluding interventions where resistance-training was performed for <1 session per week.
Meta-analyses of frequency
Meta-analyses have been performed in relation to the dose-response effect of strength training interventions, including the specific effect of frequency, by Rhea (2003) and Silva (2007). These meta-analyses investigated the dose-response effect of resistance-training (subdivided by volume, frequency and relative-load).
Rhea et al. reported that for training frequency of each muscle group, untrained individuals experience a dose-response up to 3 days per week while trained individuals experience a dose-response relationship up to 2 days per week.
Silva et al. investigated purely elderly subjects and reported that any combination of training variables led to increases in strength and only the length of time spent training had a significant dose-response relationship with strength gains.
As we will see, the findings of the meta-analysis by Rhea et al. do not entirely agree with the conclusions that we can draw from the following studies, which have directly compared lower and higher frequencies of training. This may be because the meta-analysis does not appear to control adequately for the confounding effect of volume in assessing frequency. It seems to me that what was actually assessed in that meta-analysis was the “number of training sessions” and not frequency independent of volume, which is the target of this review.
What is the effect of frequency on strength gains in trained subjects?
The following long-term training studies have explored the effects of different volume-matched frequencies of training on strength gains in trained subjects:
McLester (2000) performed a 12-week investigation involving trained subjects divided into two groups, one of which performed resistance training 1 day per week for 3 sets of upper and lower body exercises at 80% of 1RM with 2 minutes of inter-set rest. The other group trained 3 days per week for 1 set of each exercise at 80% of 1RM. The number of sets was set in order to keep total volume constant. The researchers reported the total upper and lower body exercise percentage increases in the 1-day and 3-day groups were 20.2% vs. 32.4% (upper) and 23.5% vs. 37.4% (lower). There were no significant differences between groups, although there was obviously a strong but non-significant trend for the sum of all exercise percentage increases in both the upper and lower body to be greater in the higher frequency group. Also, there was a significant difference in respect of the leg press, which displayed a 22.3% vs. 46.1% increases in favour of the high-frequency group.
Häkkinen and Kallinen (1994) performed a 6-week cross-over investigation involving trained female subjects. The subjects performed a sequence of two 3-week periods of resistance-training for the quadriceps, training 3 times a week. In one period, the subjects trained once on each training day and in the other period they trained using an identical volume over two sessions. In the 3-week period involving training once per day, the researchers observed no changes in the maximal voluntary isometric strength of the leg extensors. However, they did note significant increases in maximal isometric strength of the leg extensors of 5.1% from 2493 ± 553 to 2620 ± 598N in the 3-week period involving training two times per day. The researchers reported that this increase was much greater than that achieved in the 3-week period involving training once per day, which was an increase of just 0.1%. However, as Carpinelli (2004) has noted, these data do not match the data in the table, which report an increase of 13.2% from 2258 ± 652 to 2555 ± 555N in the 3-week period involving training once per day. It seems likely that there is an error in the data presented in the table.
Hartmann (2007) performed a 3-week investigation into the effects of twice- and once-daily training sessions with similar training volumes in 10 nationally competitive male weightlifters on isometric knee-extension strength, vertical-jump peak power and weightlifting performance. The researchers did not observe any significant differences between the two groups. However, they did find that there was a greater non-significant percentage increase in isometric knee-extension strength (5.1% vs. 3.2%) in the twice-daily training group than in the once-daily training group. It is important to note that the duration of the study very short and the training status of the subjects very high and this might have led to a greater chance of type II error occurring.
How can we summarize the literature?
The following chart summarizes the results of the studies described above:
In summary, there appears to be a trend towards a higher volume-matched frequency causing greater strength gains in trained subjects. However, there is very little evidence to build a case and further research is needed.
What is the effect of frequency on strength gains in untrained subjects?
The following long-term training studies have explored the effects of different volume-matched frequencies of training on strength gains in untrained subjects:
Calder (1994) performed a 20-week investigation in 30 young women in 3 groups who performed either whole-body training, upper-lower split training or no training (a control). The whole-body group performed 4 upper (5 sets of 6 – 10RM) and 3 lower body (5 sets of 10 – 12RM) resistance exercises in single sessions twice a week. The upper-lower split group did the upper body exercises on 2 days a week and the lower body exercises on 2 other days of the week. The researchers reported that 1RM increased significantly in the arm curl, bench press and leg press exercises in both the whole-body training and upper-lower split training groups by 54% vs. 69%, 33% vs. 32%, and 21% vs. 22%. There was therefore no difference between the strength gains attained by the two groups.
Benton (2011) investigated the effects of 8 weeks of 3 versus 4 days per week of volume-matched resistance-training on body composition in middle-aged women. The 3-day group completed 3 sets of 8 exercises arranged as a whole-body routine and the 4-day group completed 3 sets of 6 upper body exercises or 6 sets of 3 lower body exercises, arranged as an upper-lower split routine. Both groups of subjects performed 72 sets per week of 8 – 12 repetitions at 50 – 80% of 1RM. The researchers reported no significant differences in strength gains between the two groups. They found that chest press 1RM increased 34% in both groups while leg press 1RM increased 29% in the 3-day group and 49% in the 4-day group. There was therefore a trend for greater lower body strength gains in the higher frequency group.
Candow and Burke (2007) investigated the effects of 6 weeks of 2 versus 3 days per week of volume-matched resistance-training on strength gains in 29 untrained subjects, who performed either 3 sets of 10 repetitions to fatigue twice a week or 2 sets of 10 repetitions 3 times per week of the squat and bench press. The researchers reported that both groups significantly improved both squat and bench press strength. They found that the relative increases in squat 1RM for the 2-day and 3-day groups were similar (29% vs. 28%) while the relative increase in bench press 1RM was slightly higher in the higher frequency group (22% vs. 30%). However, there were no significant differences between groups.
Arazi and Asadi (2011) divided 39 healthy but untrained males into four groups: one group performing 1 session of total-body resistance training (12 exercises, once a week), another group performing total-body resistance training divided into 2 sessions (6 exercises, twice a week), an upper-lower split group performing 3 sessions per week (4 exercises, three times a week), and a control group (hereafter called 1-day, 2-day, 3-day and control groups). All groups performed the same volume and number of exercises, which comprised the leg press, leg curl, leg extension, calf raise, lat pull-down, lat pull-row, bench press, pec fly, arm curl, dumbbell arm curl, triceps push-down, and dumbbell triceps extension. Before and after the intervention, the researchers estimated bench press and leg press 1RM based on the performance of an 8RM. The researchers reported that each of the 1-day, 2-day, 3-day groups significantly increased both bench press 1RM and leg press 1RM following the intervention. However, they did not observe any significant differences in strength gains between any of the training groups. The researchers did not provide numerical figures for the improvements so it is difficult to assess whether there were any non-significant changes. However, based on the charts provided it does not appear that there were any frequency-related trends.
Hunter (1985) compared the effects of either 3-days or 4-days per week of training frequency in 46 untrained males and females. The subjects all performed 9 sets each of 7 exercises (bench press, squat, power clean, behind-the-neck press, biceps curl, behind-the-neck pull-down, and thigh curls) with a 7 – 10RM for a 7-week period. The researchers found that the 3-day and 4-day groups both significantly improved bench press strength (14.1% vs. 21.9 %) and there was no significant difference between the groups. However, the 4-day group did display a non-significantly greater improvement.
Andersen (2012) compared how distributing 1 hour per week of strength training for the neck and shoulder muscles would affect neck pain, disability and strength gains in 447 office workers with and without neck and/or shoulder pain. The subjects were randomly allocated into 1 of 4 strength training groups: 1 session of 60 minutes, 3 sessions of 20 minutes, or 9 sessions of 7 minutes, or to a non-training control group. The researchers assessed self-reported neck and shoulder pain, work disability, and strength improvements in the lateral raise exercise. The researchers reported that 10RM lateral raise performance increased by 0.16kg per week in the 1 x 60-minute group, which was significantly faster than the 9 x 7-minute group, which displayed an average increase of 0.07kg per week. The increase in the 3 x 20-minute group was 0.12kg per week but this was not significantly different from either of the other two training groups.
How can we summarize the literature?
The following chart summarizes the results of the studies displayed in the preceding section:
In summary, there is very limited evidence for the beneficial effects of either a higher volume-matched training frequency or a lower volume-matched training frequency on strength gains for untrained individuals. The research is very conflicting and it is not possible to draw a definitive conclusion at this stage.
What are the practical implications?
For trained individuals
Increasing frequency may be an effective way of maximizing strength gains, even if this occurs simply by redistributing the same volume over a greater number of sessions.
For untrained individuals
Increasing frequency may not be as effective for strength gains as in trained subjects and the research is currently conflicting. Therefore, sticking to a traditional number of sessions (e.g. three times per week) may be the most conservative course of action.